In-Depth Guide Into AIoT in 2024: When AI meets IoT

When devices we use in everyday life are getting increasingly smarter, the potential ways to utilize them are also increasing.

Due to increased interaction with people and other physical objects, IoT provides a great opportunity for collecting vast amounts of data. Artificial Intelligence (AI) applications can enable real-time analysis of the data that IoT devices have collected, helping devices to take immediate action based on the input data.

What is Artificial Intelligence of Things?

Artificial Intelligence of Things (AIoT) is a newly emerging technology that combines IoT and AI technologies to enable decision making and analytics at IoT devices:

• IoT enables networks of physical objects that are equipped with sensors, software, and other technologies to exchange data with other devices and systems over the internet
• AI enables data analytics and automated decision making

AIoT is a similar topic to edge analytics.

What are the benefits of using Artificial Intelligence of Things?

Data collected by IoT devices is useless unless it is analyzed and interpreted in a meaningful way. By adopting AI solutions in IoT, IoT devices could be enhanced by

• Real-time operational decision-making: IoT devices collect significant amounts of data. AIoT technologies enable them to use this data for real-time decision-making. A simple example is a security camera system integrated with an alarm.
  • A typical IoT camera would simply stream video data to a center where security personnel would watch the recording
  • An AIoT camera can detect trespassers and automatically activate a noise alarm to discourage the trespasser and notify the security team. Therefore, AIOT technology moves decision making from human security personnel to the IoT device, enabling labor savings and improved compliance (i.e. an automated system is better than security professionals who could fall asleep in front of the screen)
• Continually improved decisions: Devices like smart home devices collect personal preferences which can be used in improving machine learning models. Using approaches like federated learning, AIoT devices can learn from user preferences and improve their decisions.
• Reduced data transfer costs: Keeping AI systems in central locations leads to significant data transfer between edge devices and central servers. AIOT systems move analytics to edge devices and minimize data transfer.

Why Artificial Intelligence of Things is important?

AIoT enables analytics capabilities for IoT devices and as the number of IoT devices and the data that they process grows, the potential market impact of AIoT also grows.

Though different analysts have different expectations, the number of IoT-connected devices is expected to grow strongly.

Per Statista, they are expected to reach 75 billion in 2025, and per BusinessWire, they are expected to reach 41.6B and generate 79.4 zettabytes of data by 2025.

As the number of IoT devices and the data they generate is expected to grow, the main challenge in IoT will be to use this data in a meaningful way. According to a Gartner survey, 87% of organizations have limited business intelligence and analytics maturity. Therefore, most organizations underutilize their data and fail to get value from it. Integrating AI-based solutions in IoT projects will help organizations achieve better data management and analytics capabilities.

To learn more about IoT statistics, you can check our article.

What Are Potential Uses of AIoT?

Smart Cities

According the United Nations (UN);

• Currently, one in eight people lives in 33 megacities around the world
• There is expected to be 43 megacities with more than 10 million residents by 2050
• Around 70 percent of the world’s population is expected to live in urban areas by 2050

These changes are likely to bring many challenges in terms of managing cities in a sustainable and smart way. Adapting to advanced AIoT solutions can provide assistance in various areas:

• resource management (e.g. energy distribution)
• public service management
  • traffic management: Through on-street sensors and traffic light cameras, a vast array of data can be collected and analyzed to solve challenges that are faced by the public administration and city management
  • waste management

Smart Homes

Our homes are getting equipped with various devices, collecting data and providing an opportunity to utilize AIoT technologies for a better user experience. Through AIoT,

• Cost reduction and energy efficiency can be achieved by adjusting room temperature and electric usage.
• Voice assistants can leverage real-time Natural Language Processing (NLP) capabilities to improve their language understanding skills and provide customized responses

Industrial Internet of Things (IIoT)

The industrial Internet of Things (IIoT) is a technology that connects people, products, and processes to achieve digital transformation for industrial businesses. Through IIoT, complex business processes such as manufacturing can be reshaped by leveraging data. For example, AIOT systems including cameras and computer vision models can be used to visually check for quality defects.

Autonomous Things (AuT)

Autonomous Things (AuT) is a form of AIoT that performs specific tasks without human intervention. Most common AuT devices include robotics, vehicles, drones, and other autonomous software. 

• Autonomous Delivery Robots: They are mostly utilized in manufacturing, assembly, and warehousing. Using them in package delivery has already started to get common in some universities and cities and since the start of the Covid-19 outbreak, their demand is significantly increased due to the lack of human contact.
• Autonomous Vehicles: It is one of the most attention-getting AuT devices and it is estimated that 8 million devices will be shipped in 2025. For Autonomous vehicles to be successful, an AI-based computing platform, computer vision/ sensor fusion, and high-definition maps are required.

For more information on the topic, you can read our article.

Wearable Devices

Wearable devices include devices that are close to user’s skin. They detect, analyze and transmit information about body signals and provide instant feedback to the person who wears them. Smartwatches and activity trackers are examples of IoT that are becoming increasingly popular due to their convenient communication and health monitoring capabilities.

• Wearable systems for cardiac disease detection: Easy access to one-by-one health counseling became much difficult with the COVID-19 pandemic. Using smart wearable devices can mitigate that problem by remote screening and diagnosis. However, there are challenges to their adoption regarding device accuracy, clinical validity, a lack of standardized regulatory policies, and privacy concerns of patients.
• AIoT in sports science can be helpful increasing the performances of athletes and teams, as well as, sports participation and fan engagement. Data can be collected through the observation of athletes or with sensors. While some of this data can be immediately acted upon thanks to machine learning algorithms, data can also be leveraged in data visualization as well.

What are example case studies of AIOT?

ET City Brain by Alibaba Cloud

Created by Alibaba Cloud, ET City Brain is an AI-powered solution that optimizes the usage of urban public resources. The system uses big data and deep neural networks for processing logs, videos, and data streams collected from various systems and sensors in the urban center.

When it is first implemented in Hangzhou, China;

• The accuracy rate for incident identification has increased up to 92%. This helped increase accident reporting and reduce emergency response times
• Through automation of traffic signals, 3 minutes reduction in average daily commute and a 15% increase in travel speed was achieved in Xiaoshan District
• The arrival time of ambulances was reduced by half, allowing them to reach the target destination 7 minutes earlier

Autopilot by Tesla

Autopilot is an advanced driver assistance system that is utilized in Tesla’s self-driving cars. The vehicles are equipped with external cameras, ultrasonic sensors, and a powerful onboard computer to collect data. Later, the data is analyzed in a deep neural network model to determine what the car will do next. By rolling out autopilot and collecting data, Tesla is building its self-driving capabilities.

Similarly, hands-free driving is seen as an inevitable next step in the sector which will bring a safer and worry-free driving experience. With countries like Germany allowing self-driving cars on the road, this is one of the biggest areas of growth for AIoT.

What are the common vendors providing services in AIoT?

• Google
• Microsoft
• IBM
• AWS
• Oracle
• SAP
• PTC
• GE
• Salesforce
• Hitachi
• Uptake
• SAS
• Autoplant Systems
• Kairos
• Softweb Solutions
• Arundo
• C3 IoT
• Anagog
• Imagimob
• Thingstel

Further reading

If you are interested in learning more about:

• Analytics: IoT analytics
• AI
  • AI in manufacturing
  • Predictive maintenance
• IoT
  • IoT use cases
  • IoT platforms